The new era of next-generation intelligent systems is leveraging the usage of smart sensing technology to perform intelligent sensing tasks and collect useful information for different applications. This dissertation discusses secure smart sensing and applications based on the non-intrusive Photoplethysmographic (PPG) sensor, which is commonly available in current wearable devices.
In this dissertation, we first study how to authenticate a user's offline/online signature with data from the PPG sensor. We propose a novel method for offline and online signature authentication, leveraging the widely deployed PPG sensors in wrist-worn wearable devices. The unique blood flow changes in the supplicant's hand movement are being exploited in this system to validate the signature. Our experiments with real-life data sets verify the feasibility and efficiency of the proposed solutions.
In our final work, we focus on a system that can classify a user's lifted weighted object into its corresponding weight label. It leverages the change in the blood volume in the wrist region that occurred due to the strain caused by the different weights being lifted to classify the labels. We believe the importance of PPG sensing in secure smart sensing and applications during this technology era is immense.